Diabetic neuropathy (DN) is one of the most prevalent and debilitating complications of diabetes. However, DN remains largely an untreatable condition and current strategies only address pain control. A number of cellular mechanisms have been identified that lead to DN, but no explanation exists as to why certain patients develop pain, or alternatively develop insensate sensory loss. This paucity of information has hampered clinicians in developing prophylactic treatment strategies for patients that adequately prevent painful or insensate neuropathy. Our long-range goal is to understand how changes in peripheral axons can lead to varied sensory symptoms in diabetes. In previous funding years, we have identified rodent models that better reflect the varied symptoms of human DN and used these models to understand how changes in axonal subtypes within the epidermis may be related to specific symptoms of DN. Epidermal innervation is derived from two distinct axonal populations that are differentiated by different neurochemical markers and neurotrophic responsiveness. These two sensory axon subtypes are referred to as peptidergic or nonpeptidergic axons. We believe that key modifications occur in these epidermal axon subtypes early in the progression of DN that can lead to pain, or alternatively, loss of sensation. The central hypothesis is that the ratio and damage to two axonal subtypes innervating the skin is critical in specifying sensory dysfunction in diabetes. The aims of the grant include 1) testing whether changes in epidermal axons can drive pain or loss of sensation in diabetes, and testing whether inflammation and NGF responsible for changing epidermal axon phenotypes, 2) whether exercise intervention prevent or reverse changes in epidermal axon phenotypes, and 3) whether skin biopsy analyses be used to predict and follow epidermal changes associated with small fiber DN in humans. These proposed experiments are unique as they present a simple, but testable hypothesis that may explain how the divergent sensory complications can develop in diabetes. Importantly, our approaches are easily translatable to human studies and may be very relevant to other forms of neuropathy. The analyses of skin biopsies as performed here may provide new predictive power to identify patients at risk for developing neuropathy, and may also be used as a better indicator of neuropathy improvement in future clinical trials of diabetic neuropathy.